Electronic Journal of Differential Equations, Vol. 2011 (2011), No. 29,... ISSN: 1072-6691. URL: or

advertisement
Electronic Journal of Differential Equations, Vol. 2011 (2011), No. 29, pp. 1–25.
ISSN: 1072-6691. URL: http://ejde.math.txstate.edu or http://ejde.math.unt.edu
ftp ejde.math.txstate.edu
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS WITH
GIVEN PROPERTIES
MICHAL VESELÝ
Abstract. We define almost periodic functions with values in a pseudometric
space X . We mention the Bohr and the Bochner definition of almost periodicity. We present one modifiable method for constructing almost periodic
functions in X . Applying this method, we prove that in any neighbourhood
of an almost periodic skew-Hermitian linear differential system there exists a
system which does not possess a nontrivial almost periodic solution.
1. Introduction
This paper is motivated by [35] where almost periodic sequences and linear difference systems are considered. Here we will consider almost periodic functions
and linear differential systems. Our aim is to show a way one can generate almost
periodic functions with several prescribed properties. Since our process can be used
for generalizations of classical (complex valued) almost periodic functions, we introduce the almost periodicity in pseudometric spaces and we present our method
for almost periodic functions with values in a pseudometric space X .
Note that we obtain the most important case if X is a Banach space, and that the
theory of almost periodic functions of real variable with values in a Banach space,
given by Bochner [3], is in its essential lines similar to the theory of classical almost
periodic functions which is due to Bohr [4], [5]. We introduce almost periodic
functions in pseudometric spaces using a trivial extension of the Bohr concept,
where the modulus is replaced by the distance. In the classical case, we refer to the
monographs [8, 12]; for functions with values in Banach spaces, to [1], [8, Chapter
VI]; for other extensions, to [13, 18].
Necessary and sufficient conditions for a continuous function with values in a
Banach space to be almost periodic may be no longer valid for continuous functions
in general metric spaces. For the approximation condition, it is seen that the
completeness of the space of values is necessary and Tornehave [33] also requires the
local connection by arcs of the space of values. In the Bochner condition, it suffices
to replace the convergence by the Cauchy condition. Since we need the Bochner
concept as well, we recall that the Bochner condition means that any sequence
2000 Mathematics Subject Classification. 34A30, 34C27, 42A75.
Key words and phrases. Almost periodic solutions; linear differential equations;
skew-Hermitian systems.
c
2011
Texas State University - San Marcos.
Submitted August 22, 2010. Published February 18, 2011.
Supported by grant 201/09/J009 from the Czech Grant Agency.
1
2
M. VESELÝ
EJDE-2011/29
of translates of a given continuous function has a subsequence which converges,
uniformly on the domain of the function. The fact, that this condition is equivalent
with the Bohr definition of almost periodicity in Banach spaces, was proved by
Bochner [3].
The above mentioned Bohr definition and Bochner condition are formulated in
Section 2 (with some basic properties of almost periodic functions). In this section, processes from [8] are generalized. Analogously, the theory of almost periodic
functions of real variable with fuzzy real numbers as values is developed in [2]. We
remark that fuzzy real numbers form a complete metric space.
In Section 3, we mention the way one can construct almost periodic functions
with prescribed properties in a pseudometric space. We present it in Theorems 3.1,
3.2, 3.3. Note that it is possible to obtain many modifications and generalizations
of our process. A special construction of almost periodic functions with given
properties is published (and applied) in [19].
Then we will analyse almost periodic solutions of almost periodic linear differential systems. Sometimes this field is called the Favard theory what is based on
the Favard contributions in [11] (see also [6, Theorem 1.2], [12, Theorem 6.3] or
[25, Theorem 1]; for homogeneous case, see [7], [10]). In this context, sufficient
conditions for the existence of almost periodic solutions are mentioned in [16] (for
generalizations, see [15], [17]; for other extensions and supplements of the Favard
theorem, see [3], [6], [12] and the references cited therein). Certain sufficient conditions, under which homogeneous systems that have nontrivial bounded solutions
also have nontrivial almost periodic solutions, are given in [26].
It is a corollary of the Favard and the Floquet theory that any bounded solution
of an almost periodic linear differential system is almost periodic if the matrix
valued function, which determines the system, is periodic (see [12, Corollary 6.5];
for a generalization in the homogeneous case, see [14]). This result is no longer
valid for systems with almost periodic coefficients. There exist systems for which
all solutions are bounded, but none of them is almost periodic (see, e.g., [20], [27]).
Homogeneous systems have the zero solution which is almost periodic, but do not
need to have an other almost periodic solution. We note that the existence of a
homogeneous system, which has bounded solutions (separated from zero) and, at
the same time, all systems from some neighbourhood of it do not have any nontrivial
almost periodic solution, is proved in [29].
We will consider the set of systems of the form
x0 = A(t)x,
(1.1)
where A is almost periodic and all matrices A(t), t ∈ R, are skew-Hermitian, with
the uniform topology of matrix functions A on the real axis. In [28], it is proved
that the systems (1.1), all of whose solutions are almost periodic, form a dense
subset of the set of all considered systems. We add that special cases of this result
are proved in [21, 22]. For systems whose solutions are not almost periodic, we
refer to [30].
In Section 4, using the method for constructing almost periodic functions from
Section 3, we will prove that, in any neighbourhood of a system of the form (1.1),
there exists a system which does not possess an almost periodic solution other than
the trivial one, not only with a fundamental matrix which is not almost periodic
as in [30]. It means that, applying our method, we will get a stronger version of a
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
3
statement from [30]. We remark that constructions of almost periodic homogeneous
linear differential system with given properties are used in [23, 24] as well.
Let X be an arbitrary pseudometric space with a pseudometric %; i.e., let
%(x, x) = 0,
%(x, y) = %(y, x) ≥ 0,
%(x, y) ≤ %(x, z) + %(z, y)
for any x, y, z ∈ X . The symbol Oε (x) will denote the ε-neighbourhood of x in X
for arbitrary ε > 0, x ∈ X and R+
0 the set of all nonnegative real numbers. For the
used notations, we can also refer to [35].
2. Almost periodic functions in pseudometric spaces
We introduce the Bohr almost periodicity in X . Observe that we are not able to
distinguish between x ∈ X and y ∈ X if %(x, y) = 0.
Definition 2.1. A continuous function ψ : R → X is almost periodic if for any
ε > 0, there exists a number p(ε) > 0 with the property that any interval of length
p(ε) of the real line contains at least one point s, such that
%(ψ(t + s), ψ(t)) < ε,
t ∈ R.
The number s is called an ε-translation number and the set of all ε-translation
numbers of ψ is denoted by T(ψ, ε).
If X is a Banach space, then a continuous function ψ is almost periodic if and
only if any set of translates of ψ has a subsequence, uniformly convergent on R in
the sense of the norm. See, e.g., [8, Theorem 6.6]. Evidently, this result cannot be
longer valid if the space of values is not complete. Nevertheless, we prove the below
given Theorem 2.4, where the convergence is replaced by the Cauchy condition.
Before proving this statement, we mention two simple lemmas. Their proofs can be
easily obtained by modifying the proofs of [8, Theorem 6.2] and [8, Theorem 6.5],
respectively.
Lemma 2.2. An almost periodic function with values in X is uniformly continuous
on the real line.
Lemma 2.3. The set of all values of an almost periodic function ψ : R → X is
totally bounded in X .
Now we can formulate the main result of this section, which we will apply in
Section 4.
Theorem 2.4. Let ψ : R → X be a continuous function. Then, ψ is almost
periodic if and only if from any sequence of the form {ψ(t + sn )}n∈N , where sn are
real numbers, one can extract a subsequence {ψ(t + rn )}n∈N satisfying the Cauchy
uniform convergence condition on R; i.e., for any ε > 0, there exists l(ε) ∈ N with
the property that
%(ψ(t + ri ), ψ(t + rj )) < ε, t ∈ R
for all i, j > l(ε), i, j ∈ N.
Proof. The sufficiency of the condition can be proved using a simple extension of
the argument used in the proof of [8, Theorem 1.10]. In that proof, it is only
supposed, by contradiction, that any sequence of translates of ψ has a subsequence
which satisfies the Cauchy uniform convergence condition, and that ψ is not almost
periodic. Thus, it suffices to replace the modulus by the distance in the proof of [8,
Theorem 1.10].
4
M. VESELÝ
EJDE-2011/29
To prove the converse implication, we will assume that ψ is an almost periodic
function, and we will apply the well-known method of diagonal extraction and
modify the proof of [8, Theorem 6.6].
Let {tn ; n ∈ N} be a dense subset of R and {sn }n∈N ⊂ R be an arbitrarily
given sequence. From the sequence {ψ(t1 + sn )}n∈N , using Lemma 2.3, we choose
a subsequence {ψ(t1 + rn1 )}n∈N such that, for any ε > 0, there exists l1 (ε) ∈ N with
the property that
%(ψ(t1 + ri1 ), ψ(t1 + rj1 )) < ε,
i, j > l1 (ε), i, j ∈ N.
Such a subsequence exists because infinitely many values ψ(t1 + sn ) is in a neighbourhood of radius 2−i for all i ∈ N (consider the method of diagonal extraction).
Analogously, from the sequence {ψ(t2 + rn1 )}n∈N , we obtain {ψ(t2 + rn2 )}n∈N such
that, for any ε > 0, there exists l2 (ε) ∈ N for which
%(ψ(t2 + ri2 ), ψ(t2 + rj2 )) < ε,
i, j > l2 (ε), i, j ∈ N.
We proceed further in the same way. We obtain {rnk } ⊆ · · · ⊆ {rn1 }, k ∈ N.
Let ε > 0 be arbitrarily given, p = p(ε/5) be from Definition 2.1, δ = δ(ε/5)
correspond to ε/5 from the definition of the uniform continuity of ψ (see Lemma
2.2) and let a finite set {t1 , . . . , tj } ⊂ {tn ; n ∈ N} satisfy
min
i∈{1,...,j}
|ti − t| < δ,
t ∈ [0, p].
Obviously, there exists l ∈ N such that, for all integers n1 , n2 > l, it is valid
ε
%(ψ(ti + rnn11 ), ψ(ti + rnn22 )) < , i ∈ {1, . . . , j}.
5
Let t ∈ R be given, s = s(t) ∈ [−t, −t + p] be an ε/5-translation number of ψ, and
ti = ti (s) ∈ {t1 , . . . , tj } be such that |t + s − ti | < δ. Finally, we have
%(ψ(t + rnn11 ), ψ(t + rnn22 ))
≤ %(ψ(t + rnn11 ), ψ(t + rnn11 + s)) + %(ψ(t + rnn11 + s), ψ(ti + rnn11 ))
+ %(ψ(ti + rnn11 ), ψ(ti + rnn22 )) + %(ψ(ti + rnn22 ), ψ(t + rnn22 + s))
+ %(ψ(t + rnn22 + s), ψ(t + rnn22 )).
Thus, we obtain
ε ε ε ε ε
+ + + + =ε
(2.1)
5 5 5 5 5
for all t ∈ R, n1 , n2 > l, n1 , n2 ∈ N. Evidently, (2.1) completes the proof of the
theorem if we put rn := rnn , n ∈ N.
%(ψ(t + rnn11 ), ψ(t + rnn22 )) <
Analogously as for almost periodic functions with values in a Banach space, one
can prove many properties of almost periodic functions in a pseudometric space. For
example, the limit of a uniformly convergent sequence of almost periodic functions
is almost periodic (see [8, Theorem 6.4]).
Using Theorem 2.4 n-times, one can also obtain:
Corollary 2.5. If X1 , . . . , Xn are pseudometric spaces and ψ1 , . . . , ψn are arbitrary almost periodic functions with values in X1 , . . . , Xn , respectively, then the
function ψ, with values in X1 × · · · × Xn given by ψ := (ϕ1 , . . . , ϕn ), is almost
periodic as well.
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
5
Moreover, from Corollary 2.5 it follows that the set
T(ψ1 , ε) ∩ T(ψ2 , ε) ∩ · · · ∩ T(ψn , ε)
is relative dense in R for arbitrary almost periodic functions ψ1 , ψ2 , . . . , ψn and any
ε > 0. We add that one can use Corollary 2.5 to obtain simple modifications of the
below presented method of constructions of almost periodic functions.
3. Construction of almost periodic functions
Now we present the way one can generate almost periodic functions with given
properties in the next theorem.
Theorem 3.1. For arbitrary a > 0, any continuous function ψ : R → X such that
ψ(t) ∈ Oa (ψ(t − 1)),
ψ(t) ∈ Oa (ψ(t + 2)),
ψ(t) ∈ Oa/2 (ψ(t − 4)),
ψ(t) ∈ Oa/2 (ψ(t + 8)),
ψ(t) ∈ Oa/4 (ψ(t − 24 )),
ψ(t) ∈ Oa/4 (ψ(t + 25 )),
t ∈ (1, 2],
t ∈ (−2, 0],
t ∈ (2, 6],
t ∈ (−10, −2],
t ∈ (2 + 22 , 2 + 22 + 24 ],
t ∈ (−25 − 23 − 2, −23 − 2],
...
2n
ψ(t) ∈ Oa 2−n (ψ(t − 2 )),
2
t ∈ (2 + 2 + · · · + 22n−2 , 2 + 22 + · · · + 22n−2 + 22n ],
ψ(t) ∈ Oa 2−n (ψ(t + 22n+1 )),
t ∈ (−22n+1 − · · · − 23 − 2, −22n−1 − · · · − 23 − 2],
...
is almost periodic.
Proof. Let ε > 0 be arbitrary and k = k(ε) ∈ N be such that 2k > 8a/ε. We have
to prove that the set of all ε-translation numbers of ψ is relative dense in R. We
will obtain this from the fact that l · 22k is an ε-translation number of ψ for any
integer l. We see that
ψ(t) ∈ Oε/8 (ψ(t − 22k )),
t ∈ (2 + 22 + · · · + 22k−2 , 2 + 22 + · · · + 22k ],
ψ(t) ∈ Oε/8 (ψ(t + 22k+1 )), t ∈ (−22k+1 − · · · − 23 − 2, −22k−1 − · · · − 23 − 2],
ψ(t) ∈ Oε/16 (ψ(t − 22k+2 )),
t ∈ (2 + 22 + · · · + 22k , 2 + 22 + · · · + 22k+2 ],
...
In a pseudometric space X , this implies
ψ(t + 22k ) ∈ Oε/8 (ψ(t)),
t ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ],
ψ(t − 22k+1 ) ∈ Oε/8 (ψ(t)),
ψ(t − 22k ) ∈ Oε/8+ε/8 (ψ(t)),
ψ(t + 22k+1 ) ∈ Oε/8+ε/16 (ψ(t)),
t ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ],
t ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ],
t ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ],
ψ(t + 3 · 22k ) ∈ Oε/8+ε/8+ε/16 (ψ(t)),
t ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ],
ψ(t + 22k+2 ) ∈ Oε/16 (ψ(t)),
t ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ],
6
M. VESELÝ
EJDE-2011/29
ψ(t + 22k + 22k+2 ) ∈ Oε/8+ε/16 (ψ(t)),
t ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ],
...
Since
ε ε
ε
ε
ε
ε
+ +
+
+
+ ··· = ,
8 8 16 16 32
2
we have
ψ(t + l · 22k ) ∈ Oε/2 (ψ(t)),
t ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ], l ∈ Z.
We express any t ∈ R as the sum of numbers p(t) and q(t) for which
(3.1)
p(t) ∈ [−22k−1 − · · · − 23 − 2, 2 + 22 + · · · + 22k−2 ],
q(t) ∈ Z and q(t) = j22k
for some j ∈ Z.
Using (3.1), we obtain
%(ψ(t), ψ(t + l · 22k ))
≤ %(ψ(p(t) + q(t)), ψ(p(t))) + %(ψ(p(t)), ψ(p(t) + (j + l) 22k )) <
for any t ∈ R, l ∈ Z, which completes the proof.
(3.2)
ε ε
+ =ε
2 2
The process mentioned in the previous theorem is easily modifiable. We illustrate
this fact by the following two theorems. We remark that Theorem 3.2 is used in
[34].
Theorem 3.2. Let M > 0, x0 ∈ X , and j ∈ N be given. Let ϕ : [0, M ] → X satisfy
ϕ(0) = ϕ(M ) = x0 . If {rn }n∈N ⊂ R+
0 has the property that
∞
X
rn < ∞,
(3.3)
n=1
then an arbitrary continuous function ψ : R → X , ψ|[0,M ] ≡ ϕ for which
ψ(t) = x0 ,
t ∈ {iM, 2 ≤ i ≤ j + 1} ∪ {−i(j + 1)M, 1 ≤ i ≤ j}∪
2n−2
∪∞
+ i(j + 1)2n )M ; 1 ≤ i ≤ j}∪
n=1 {((j + 1) + · · · + j(j + 1)
2n−1
∪∞
+ i(j + 1)2n+1 )M ; 1 ≤ i ≤ j}
n=1 {−((j + 1) + · · · + j(j + 1)
and, at the same time, for which it is valid
ψ(t) ∈ Or1 (ψ(t − M )),
t ∈ (M, 2M ),
...
ψ(t) ∈ Or1 (ψ(t − jM )),
t ∈ (jM, (j + 1)M ),
ψ(t) ∈ Or2 (ψ(t + (j + 1)M )),
t ∈ (−(j + 1)M, 0),
...
ψ(t) ∈ Or2 (ψ(t + j(j + 1)M )),
t ∈ (−j(j + 1)M, −(j − 1)(j + 1)M ),
2
t ∈ ((j + 1)M, ((j + 1) + (j + 1)2 )M ),
ψ(t) ∈ Or3 (ψ(t − (j + 1) M )),
...
2
ψ(t) ∈ Or3 (ψ(t − j(j + 1) M )),
t ∈ (((j + 1) + (j − 1)(j + 1)2 )M, ((j + 1) + j(j + 1)2 )M ),
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
7
...
2n−1
ψ(t) ∈ Or2n (ψ(t + (j + 1)
2n−1
M )),
+ · · · + j(j + 1)3 + j(j + 1))M,
− (j(j + 1)2n−3 + · · · + j(j + 1)3 + j(j + 1))M ,
t ∈ −((j + 1)
2n−3
+ j(j + 1)
...
2n−1
ψ(t) ∈ Or2n (ψ(t + j(j + 1)
M )),
t ∈ −(j(j + 1)2n−1 + j(j + 1)2n−3 + · · · + j(j + 1)3 + j(j + 1))M,
− ((j − 1)(j + 1)2n−1 + j(j + 1)2n−3 + · · · + j(j + 1)3 + j(j + 1))M ,
ψ(t) ∈ Or2n+1 (ψ(t − (j + 1)2n M )),
t ∈ ((j + 1) + j(j + 1)2 + · · · + j(j + 1)2n−2 )M,
((j + 1) + j(j + 1)2 + · · · + j(j + 1)2n−2 + (j + 1)2n )M ,
...
2n
ψ(t) ∈ Or2n+1 (ψ(t − j(j + 1) M )),
t ∈ ((j + 1) + j(j + 1)2 + · · · + j(j + 1)2n−2 + (j − 1)(j + 1)2n )M,
((j + 1) + j(j + 1)2 + · · · + j(j + 1)2n−2 + j(j + 1)2n )M ,
...
is almost periodic.
Proof. We can prove this theorem analogously as Theorem 3.1. Let ε be a positive
number and let an odd integer n(ε) ≥ 2 have the property (see (3.3)) that
∞
X
rn <
n=n(ε)
ε
.
2
(3.4)
We will prove that l(j + 1)n(ε)−1 M is an ε-translation number of ψ for all l ∈ Z.
Arbitrarily choosing l ∈ Z and t ∈ R, if we put
s := l(j + 1)n(ε)−1 M,
(3.5)
then it suffices to show that the inequality
d(ψ(t), ψ(t + s)) < ε
(3.6)
holds; i.e., this inequality proves the theorem.
We can write t as the sum of numbers t1 and t2 , where
t1 ≥ − j(j + 1)n(ε)−2 + · · · + j(j + 1)3 + j(j + 1) M,
t1 ≤ j + 1 + j(j + 1)2 + · · · + j(j + 1)n(ε)−3 M
(3.7)
and
t2 = i(j + 1)n(ε)−1 M for some i ∈ Z.
Now we have (see (3.7) and the proof of Theorem 3.1)
(3.8)
%(ψ(t), ψ(t + s)) ≤ %(ψ(t1 + t2 ), ψ(t1 )) + %(ψ(t1 ), ψ(t1 + t2 + s))
n(ε)+p−1
<
X
n=n(ε)
n(ε)+q−1
rn +
X
n=n(ε)
rn .
(3.9)
8
M. VESELÝ
EJDE-2011/29
Indeed, we can express (consider (3.5) and (3.8))
t2 = i1 (j + 1)n(ε)−1 + i2 (j + 1)n(ε) + · · · + ip (j + 1)n(ε)+p−1 (m + 1),
t2 + s = l1 (j + 1)n(ε)−1 + l2 (j + 1)n(ε) + · · · + lq (j + 1)n(ε)+q−1 (m + 1),
where i1 , . . . , ip , l1 , . . . , lq ⊆ {−j, . . . , 0, . . . , j} satisfy
i1 ≥ 0, i2 ≤ 0, . . . , (−1)p ip ≤ 0,
l1 ≥ 0, l2 ≤ 0, . . . , (−1)q lq ≤ 0,
and use (iii). It is sure that (3.4) and (3.9) give (3.6).
Theorem 3.3. Let ϕ : (−r, r] → X , {rn }n∈N ⊂
such that
∞
X
rn jn < ∞
R+
0,
and {jn }n∈N ⊆ N be arbitrary
(3.10)
n=1
holds, and let a function ψ : R → X satisfy ψ|(−r,r] ≡ ϕ and
ψ(t) ∈ Or1 (ϕ(t − 2r)),
...
ψ(t) ∈ Or1 (ϕ(t − 2r)),
t ∈ (r + (j1 − 1)2r, r + j1 2r],
ψ(t) ∈ Or1 (ϕ(t + 2r)),
...
ψ(t) ∈ Or1 (ϕ(t + 2r)),
t ∈ (r, r + 2r],
t ∈ (−2r − r, −r],
t ∈ (−j1 2r − r, −(j1 − 1)2r − r],
...
ψ(t) ∈ Orn (ϕ(t − pn )),
t ∈ (p1 + · · · + pn−1 , p1 + · · · + pn−1 + pn ],
...
ψ(t) ∈ Orn (ϕ(t − pn )),
t ∈ (p1 + · · · + pn−1 + (jn − 1)pn , p1 + · · · + pn−1 + jn pn ],
ψ(t) ∈ Orn (ϕ(t + pn )),
t ∈ (−pn − pn−1 − · · · − p1 , −pn−1 − · · · − p1 ],
...
ψ(t) ∈ Orn (ϕ(t + pn )),
t ∈ (−jn pn − pn−1 − · · · − p1 , −(jn − 1)pn − pn−1 − · · · − p1 ],
...
where
p1 := r + j1 2r,
p3 := (2j2 + 1)p2 ,
...,
p2 := 2(r + j1 2r),
pn := (2jn−1 + 1)pn−1 , . . .
If ψ is continuous on R, then it is almost periodic.
Proof. It is not difficult to prove Theorem 3.3 analogously as Theorems 3.1 and
3.2. For given ε > 0, let an integer n(ε) ≥ 2 satisfy
∞
X
n=n(ε)
rn jn <
ε
.
4
One can prove the inclusion
{lpn(ε) ; l ∈ Z} ⊆ T(ψ, ε)
(3.11)
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
which yields the almost periodicity of ψ.
9
Remark 3.4. From the proofs of Theorems 3.1, 3.2, 3.3 (see (3.2), (3.5) and (3.6),
(3.11)), we obtain a property of the set of all ε-translation numbers of the resulting
function ψ. For any ε > 0, there exists nonzero c ∈ R for which
{lc; l ∈ Z} ⊆ T(ψ, ε).
An important class of almost periodic functions is the class of limit-periodic functions. To this class belong the uniform limits of sequences of periodic continuous
functions (in general, having different periods). It is seen that, applying the method
from the above theorems, we obtain limit-periodic functions.
4. An application
Let m ∈ N be arbitrarily given. In this section, we will use the following notations: Im(ϕ) for the range of a function ϕ, Mat(C, m) for the set of all m × m
matrices with complex elements, U (m) ⊂ Mat(C, m) for the group of all unitary
matrices of dimension m, A∗ for the conjugate transpose of A ∈ Mat(C, m), I for
the identity matrix, 0 for the zero matrix, and i for the imaginary unit.
We will analyse systems of m homogeneous linear differential equations of the
form
x0 (t) = A(t)x(t), t ∈ R,
(4.1)
where A is an almost periodic function with Im(A) ⊂ Mat(C, m) and with the
property that A(t) + A∗ (t) = 0 for any t ∈ R; i.e., A : R → Mat(C, m) is an almost
periodic function of skew-Hermitian (skew-adjoint) matrices. Let S be the set of
all systems (4.1). We will identify the function A with the system (4.1) which is
determined by A. Especially, we will write A ∈ S.
In the vector space Cm , we will consider the absolute norm k · k1 (one can also
consider the Euclidean norm or the maximum norm). Let k · k be the corresponding
matrix norm. Considering that every almost periodic function is bounded (see
Lemma 2.3), the distance between two systems A, B ∈ S is defined by the norm of
the matrix valued functions A, B, uniformly on R; i.e., we introduce the metric
σ(A, B) := sup kA(t) − B(t)k,
A, B ∈ S.
t∈R
For ε > 0, the symbol Oεσ (A) will denote the ε-neighbourhood of A in S.
Now we recall the notion of the frequency module and its rational hull which
can be introduced for all almost periodic function with values in a Banach space.
The frequency module F of an almost periodic function A : R → Mat(C, m) is the
Z-module of the real numbers, generated by the λ such that
Z
1 T
lim
A(t)e2πiλt dt 6= 0.
T →+∞ T 0
The rational hull of F is the set
{λ/l; λ ∈ F, l ∈ Z}.
For the frequency modules of almost periodic linear differential systems and their
solutions, we refer to [12, Chapters 4, 6], [25].
In [28], it is proved that, in any neighbourhood of a system (4.1) with frequency
module F, there exists a system with a frequency module contained in the rational
hull of F possessing all almost periodic solutions with frequencies belonging to
10
M. VESELÝ
EJDE-2011/29
the rational hull of F as well. From [31, Theorem 1] it follows that there exists
a system (4.1) which cannot be approximated by the so-called reducible systems
with frequency module F (there exists an open set of irreducible systems with a
fixed frequency module); i.e., a neighbourhood of a system (4.1) may not contain
a system with almost periodic solutions and frequency module F. In this case, see
also [9] and [32] for reducible constant systems and systems reducing to diagonal
form by Lyapunov transformation with frequency module F, respectively.
In addition, it is proved in [30] that the systems with k-dimensional frequency
basis of A, having solutions which are not almost periodic, form a subset of the
second category of the space of all considered systems with k-dimensional frequency
basis of A. Thus, it is known (see [28, Corollary 1]) that the systems with k-dimensional frequency basis of A and with an almost periodic fundamental matrix form
a dense set of the first category in the space of all systems (4.1) with k-dimensional
frequency basis.
In this context, we formulate the following result that the systems having no
nontrivial almost periodic solution form a dense subset of S.
Theorem 4.1. For any A ∈ S and ε > 0, there exists B ∈ Oεσ (A) which does not
have an almost periodic solution other than the trivial one.
Proof. Let A, C ∈ S and ε > 0 be arbitrary. Since the sum of skew-Hermitian
matrices is also skew-Hermitian and since the sum of two almost periodic functions
is almost periodic (consider Theorem 2.4), we have that A+C ∈ S. Let XA (t), t ∈ R
and XC (t), t ∈ R be the principal (i.e., XA (0) = XC (0) = I) fundamental matrices
of A ∈ S and C ∈ S, respectively. If the matrices C(t), XA (t) commute for all t ∈ R,
then the matrix valued function XA (t) · XC (t), t ∈ R is the principal fundamental
0
(t) = A(t) · XA (t), XC0 (t) = C(t) · XC (t) for
matrix of A + C ∈ S. Indeed, from XA
t ∈ R, we obtain
(XA (t)XC (t))0 = A(t)XA (t)XC (t) + XA (t)C(t)XC (t)
= A(t)XA (t)XC (t) + C(t)XA (t)XC (t)
= (A + C)(t)XA (t)XC (t),
t ∈ R.
It gives that it suffices to find C ∈ Oεσ (0) for which all matrices C(t), t ∈ R have the
form diag[ia, . . . , ia], a ∈ R and for which the vector valued function XA (t)·XC (t)·u,
t ∈ R is not almost periodic for any vector u ∈ Cm , kuk1 = 1.
We will construct such an almost periodic function C applying Theorem 3.1 for
a = ε/4. First of all we put
C(t) ≡ 0,
t ∈ [0, 1].
Then, in the first step of our construction, we define C on (1, 2] arbitrarily so
that it is constant on [1 + 1/4, 1 + 3/4] and kC(t)k < ε/4 for t from this interval,
C(2) := C(1) = 0, and it is linear between values 0, C(3/2) on [1, 1 + 1/4] and
[1 + 3/4, 2].
In the second step, we define continuous C satisfying kC(t) − C(t + 2)k < ε/4
for t ∈ [−2, 0) arbitrarily so that it is constant on
[−2 + 1/16, −2 + 1 − 1/16],
[−2 + 1 + 1/4 + 1/16, −2 + 1 + 3/4 − 1/16];
at the same time, we put
C(−2) := C(0) = 0,
C(−1 + 1/4) := C(1 + 1/4) = C(3/2),
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
C(−1/4) := C(2 − 1/4) = C(3/2),
C(−1) := C(1) = 0,
C(t) ≡ C(3/2)/2,
11
t ∈ [−1 + 1/16, −1 + 1/4 − 1/16] ∪ [−1/4 + 1/16, −1/16]
and define C so that it is linear on
[−2, −2 + 1/16],
[−1 − 1/16, −1],
[−1 + 1/4 − 1/16, −1 + 1/4],
[−1/4 − 1/16, −1/4],
[−1, −1 + 1/16],
[−1 + 1/4, −1 + 1/4 + 1/16],
[−1/4, −1/4 + 1/16],
[−1/16, 0].
Analogously, in the third step, we obtain C on (2, 6] for which we can choose
constant values on
[4 − 2 + 1/16 + 8−1 /16, 4 − 2 + 1 − 1/16 − 8−1 /16],
[4 − 2 + 1 + 1/4 + 1/16 + 8−1 /16, 4 − 2 + 1 + 3/4 − 1/16 − 8−1 /16],
[4 − 1 + 1/16 + 8−1 /16, 4 − 1 + 1/4 − 1/16 − 8−1 /16],
[4 − 1/4 + 1/16 + 8−1 /16, 4 − 1/16 − 8−1 /16],
[4 + 8−1 /16, 4 + 1 − 8−1 /16],
[4 + 1 + 1/4 + 8−1 /16, 4 + 1 + 3/4 − 8−1 /16]
arbitrarily so that kC(t) − C(t − 4)k < ε/8, t ∈ (2, 6]; at the same time, we put
C(4 − 2 + 1/16) := C(−2 + 1/16) = C(−3/2),
C(4 − 2 + 1 − 1/16) := C(−1 − 1/16) = C(−3/2),
C(4 − 1) := C(−1) = 0,
C(4 − 1 + 1/16) := C(−1 + 1/16) = C(3/2)/2,
C(4 − 1 + 1/4 − 1/16) := C(−1 + 1/4 − 1/16) = C(3/2)/2,
C(4 − 1 + 1/4) := C(−1 + 1/4) = C(3/2),
C(4 − 2 + 1 + 1/4 + 1/16) := C(−2 + 1 + 1/4 + 1/16) = C(−1/2),
C(4 − 2 + 1 + 3/4 − 1/16) := C(−2 + 1 + 3/4 − 1/16) = C(−1/2),
C(4 − 1/4) := C(−1/4) = C(3/2),
C(4 − 1/4 + 1/16) := C(−1/4 + 1/16) = C(3/2)/2,
C(4 − 1/16) := C(−1/16) = C(3/2)/2,
C(4) := C(0),
C(4 + 1) := C(1),
C(4 + 1 + 1/4) := C(1 + 1/4) = C(3/2),
C(4 + 1 + 3/4) := C(1 + 3/4) = C(3/2),
C(4 + 2) := C(2) = C(0) = 0,
C(t) ≡ C(−3/2)/2,
t ∈ [4 − 2 + 8−1 /16, 4 − 2 + 1/16 − 8−1 /16]
∪ [4 − 1 − 1/16 + 8−1 /16, 4 − 1 − 8−1 /16],
C(t) ≡ C(3/2)/4,
t ∈ [4 − 1 + 8−1 /16, 4 − 1 + 1/16 − 8−1 /16]
∪ [4 − 1/16 + 8−1 /16, 4 − 8−1 /16],
C(t) ≡ 3 C(3/2)/4,
t ∈ [4 − 1 + 1/4 − 1/16 + 8−1 /16, 4 − 1 + 1/4 − 8−1 /16]
∪ [4 − 1/4 + 8−1 /16, 4 − 1/4 + 1/16 − 8−1 /16],
12
M. VESELÝ
EJDE-2011/29
C(t) ≡(C(3/2) + C(−1/2))/2,
t ∈ [4 − 1 + 1/4 + 8−1 /16, 4 − 1 + 1/4 + 1/16 − 8−1 /16]
∪ [4 − 1/4 − 1/16 + 8−1 /16, 4 − 1/4 − 8−1 /16],
C(t) ≡(8 C(4 + 1) + 1 C(4 + 1 + 1/4))/9,
t ∈ [4 + 1 + 8−1 /16, 4 + 1 + 8−1 /16 · 3],
C(t) ≡ (7 C(4 + 1) + 2 C(4 + 1 + 1/4))/9,
t ∈ [4 + 1 + 8−1 /16 · 5, 4 + 1 + 8−1 /16 · 7],
...
C(t) ≡(1 C(4 + 1) + 8 C(4 + 1 + 1/4))/9,
t ∈ [4 + 1 + 8−1 /16 · 29, 4 + 1 + 8−1 /16 · 31],
C(t) ≡(8 C(4 + 1 + 3/4) + 1 C(4 + 2))/9,
t ∈ [4 + 1 + 3/4 + 8−1 /16, 4 + 1 + 3/4 + 8−1 /16 · 3],
C(t) ≡ (7 C(4 + 1 + 3/4) + 2 C(4 + 2))/9,
t ∈ [4 + 1 + 3/4 + 8−1 /16 · 5, 4 + 1 + 3/4 + 8−1 /16 · 7],
...
C(t) ≡ (1 C(4 + 1 + 3/4) + 8 C(4 + 2))/9,
t ∈ [4 + 1 + 3/4 + 8−1 /16 · 29, 4 + 1 + 3/4 + 8−1 /16 · 31] .
Then we define continuous C so that it is linear on the rest of subintervals.
If we denote
a11 := 0,
a12 := 1,
b11 := 0,
b12 := 1 + 1/4,
c11 := 1,
c12 := 1 + 3/4,
a13 := 2
and (compare with the situation after the second step)
a21 := −2,
a22 := −2,
b22 := −2 + 1/16,
a23 := −1,
a24 := −1,
a25 := −1 + 1/4,
b21 := −2,
b23 := −1,
b24 := −1 + 1/16,
c21 := −2,
c22 := −1 − 1/16,
c23 := −1,
c24 := −1 + 1/4 − 1/16,
b25 := −1 + 1/4 + 1/16,
a26 := −1 + 3/4,
c25 := −1 + 3/4 − 1/16,
b26 := −1 + 3/4 + 1/16,
c26 := −1/16,
we see that C does not need to be constant only on
[a1j − 2, a1j − 2 + 4−2 ],
[c1j − 2 − 4−2 , c1j − 2],
[b12 − 2 − 4−2 , b12 − 2],
[c1j − 2, c1j − 2 + 4−2 ],
[b1j − 2, b1j − 2 + 4−2 ],
[a1j+1 − 2 − 4−2 , a1j+1 − 2]
for j ∈ {1, 2}; i.e., on
[a2j , b2j ], j ∈ {1, . . . , 6},
[c2j , a2j+1 ], j ∈ {1, . . . , 5},
[c26 , 0],
and it has to be constant on each one of the intervals
[a12 − 2 + 4−2 , b12 − 2 − 4−2 ],
[c12 − 2 + 4−2 , a13 − 2 − 4−2 ],
[b1j − 2 + 4−2 , c1j − 2 − 4−2 ],
j ∈ {1, 2},
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
i.e., on [b2j , c2j ], j ∈ {1, . . . , 6}. It is also seen that
a21 = d11 ,
b21 = d12 ,
c21 = d13 ,
a22 = d14 ,
...
c26 = d118 ,
where d11 , d12 , . . . , d118 is the nondecreasing sequence of all numbers
a1j − 2,
b1j − 2,
c1j − 2,
min{a1j − 2 + 4−2 , b1j − 2},
max{a1j − 2, b1j − 2 − 4−2 },
min{c1j − 2, b1j − 2 + 4−2 },
max{c1j − 2 − 4−2 , b1j − 2},
min{c1j − 2 + 4−2 , a1j+1 − 2},
max{c1j − 2, a1j+1 − 2 − 4−2 }
for j ∈ {1, 2}. We put a27 := 0.
Let d21 , d22 , . . . , d2168 be the nondecreasing sequence of all numbers
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
b11 + 4,
c11 + 4,
min{c11 + 4, b11 + 4 + 8−1 /16},
max{c11 + 4 − 8−1 /16, b11 + 4},
c12 + 4,
min{c12 + 4, b12 + 4 + 8−1 /16},
max{c12 + 4 − 8−1 /16, b12 + 4},
a11 + (4k + 1)(b11 − a11 )/32 + 4,
a11 + (4k + 3)(b11 − a11 )/32 + 4,
a11 + (4k + 4)(b11 − a11 )/32 + 4,
c11
+
a12 +
c12 +
k ∈ {0, 1, . . . , 7},
1)(a12
(4k +
− c11 )/32 + 4, c11 + (4k + 3)(a12 − c11 )/32 + 4,
c11 + (4k + 4)(a12 − c11 )/32 + 4, k ∈ {0, 1, . . . , 7},
(4k + 1)(b12 − a12 )/32 + 4, a12 + (4k + 3)(b12 − a12 )/32 + 4,
a12 + (4k + 4)(b12 − a12 )/32 + 4, k ∈ {0, 1, . . . , 7},
(4k + 1)(a13 − c12 )/32 + 4, c12 + (4k + 3)(a13 − c12 )/32 + 4,
c12 + (4k + 4)(a13 − c12 )/32 + 4, k ∈ {0, 1, . . . , 7}
and
a2j+1 + 4,
b2j + 4,
c2j + 4,
min{a2j + 4 + 8−1 /16, b2j + 4},
max{a2j + 4, b2j + 4 − 8−1 /16},
min{c2j + 4, b2j + 4 + 8−1 /16},
max{c2j + 4 − 8−1 /16, b2j + 4},
min{c2j + 4 + 8−1 /16, a2j+1 + 4},
max{c2j + 4, a2j+1 + 4 − 8−1 /16}
for j ∈ {1, . . . , 6}. We denote
a31 := 2,
b31 := d21 ,
c31 := d22 ,
a32 := d23 ,
...
a357 := d2168 .
We remark that, in the sequences of dlj , l ∈ N, values are a number of time.
In the fourth step, we define C so that
ε
kC(t) − C(t + 23 )k < 3 , t ∈ [−23 − 2, −2).
2
We consider the nondecreasing sequence d31 , d32 , . . . , d321·82 of values
a3j − 23 ,
b3j − 23 ,
c3j − 23 ,
min{a3j − 23 + 8−2 /16, b3j − 23 },
max{a3j − 23 , b3j − 23 − 8−2 /16},
min{c3j − 23 , b3j − 23 + 8−2 /16},
max{c3j − 23 − 8−2 /16, b3j − 23 },
13
14
M. VESELÝ
min{c3j − 23 + 8−2 /16, a3j+1 − 23 },
EJDE-2011/29
max{c3j − 23 , a3j+1 − 23 − 8−2 /16}
for j ∈ {1, . . . , 7 · 8}, 144 numbers b11 − 23 , and
c11 − 23 ,
min{c11 − 23 , b11 − 23 + 8−2 /16},
max{c11 − 23 − 8−2 /16, b11 − 23 },
c12 − 23 ,
min{c12 − 23 , b12 − 23 + 8−2 /16},
max{c12 − 23 − 8−2 /16, b12 − 23 },
min{a11 + (k − 1)(b11 − a11 )/(8 · 4) − 23 + 8−2 /16, a11 + k(b11 − a11 )/(8 · 4) − 23 },
max{a11 + (k − 1)(b11 − a11 )/(8 · 4) − 23 , a11 + k(b11 − a11 )/(8 · 4) − 23 − 8−2 /16},
a11 + k(b11 − a11 )/(8 · 4) − 23 ,
k ∈ {1, . . . , 8 · 4},
min{c11 + (k − 1)(a12 − c11 )/(8 · 4) − 23 + 8−2 /16, c11 + k(a12 − c11 )/(8 · 4) − 23 },
max{c11 + (k − 1)(a12 − c11 )/(8 · 4) − 23 , c11 + k(a12 − c11 )/(8 · 4) − 23 − 8−2 /16},
c11 + k(a12 − c11 )/(8 · 4) − 23 , k ∈ {1, . . . , 8 · 4},
min{a12 + (k − 1)(b12 − a12 )/(8 · 4) − 23 + 8−2 /16, a12 + k(b12 − a12 )/(8 · 4) − 23 },
max{a12 + (k − 1)(b12 − a12 )/(8 · 4) − 23 , a12 + k(b12 − a12 )/(8 · 4) − 23 − 8−2 /16},
a12 + k(b12 − a12 )/(8 · 4) − 23 , k ∈ {1, . . . , 8 · 4},
min{c12 + (k − 1)(a13 − c12 )/(8 · 4) − 23 + 8−2 /16, c12 + k(a13 − c12 )/(8 · 4) − 23 },
max{c12 + (k − 1)(a13 − c12 )/(8 · 4) − 23 , c12 + k(a13 − c12 )/(8 · 4) − 23 − 8−2 /16},
c12 + k(a13 − c12 )/(8 · 4) − 23 , k ∈ {1, . . . , 8 · 4},
c21 − 23 , min{c21 − 23 , b21 − 23 + 8−2 /16}, max{c21 − 23 − 8−2 /16, b21 − 23 },
c22 − 23 , min{c22 − 23 , b22 − 23 + 8−2 /16}, max{c22 − 23 − 8−2 /16, b22 − 23 },
c23 − 23 , min{c23 − 23 , b23 − 23 + 8−2 /16}, max{c23 − 23 − 8−2 /16, b23 − 23 },
c24 − 23 , min{c24 − 23 , b24 − 23 + 8−2 /16}, max{c24 − 23 − 8−2 /16, b24 − 23 },
c25 − 23 , min{c25 − 23 , b25 − 23 + 8−2 /16}, max{c25 − 23 − 8−2 /16, b25 − 23 },
c26 − 23 , min{c26 − 23 , b26 − 23 + 8−2 /16}, max{c26 − 23 − 8−2 /16, b26 − 23 },
min{a21 + (k − 1)(b21 − a21 )/8 − 23 + 8−2 /16, a21 + k(b21 − a21 )/8 − 23 },
max{a21 + (k − 1)(b21 − a21 )/8 − 23 , a21 + k(b21 − a21 )/8 − 23 − 8−2 /16},
a21 + k(b21 − a21 )/8 − 23 , k ∈ {1, . . . , 8},
min{c21 + (k − 1)(a22 − c21 )/8 − 23 + 8−2 /16, c21 + k(a22 − c21 )/8 − 23 },
max{c21 + (k − 1)(a22 − c21 )/8 − 23 , c21 + k(a22 − c21 )/8 − 23 − 8−2 /16},
c21 + k(a22 − c21 )/8 − 23 , k ∈ {1, . . . , 8},
...
min{a26 + (k − 1)(b26 − a26 )/8 − 23 + 8−2 /16, a26 + k(b26 − a26 )/8 − 23 },
max{a26 + (k − 1)(b26 − a26 )/8 − 23 , a26 + k(b26 − a26 )/8 − 23 − 8−2 /16},
a26 + k(b26 − a26 )/8 − 23 , k ∈ {1, . . . , 8},
min{c26 + (k − 1)(a27 − c26 )/8 − 23 + 8−2 /16, c26 + k(a27 − c26 )/8 − 23 },
max{c26 + (k − 1)(a27 − c26 )/8 − 23 , c26 + k(a27 − c26 )/8 − 23 − 8−2 /16},
c26 + k(a27 − c26 )/8 − 23 , k ∈ {1, . . . , 8}.
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
15
We put
a41 := d31 ,
b41 := d32 ,
c41 := d33 ,
...
c47·82 := d321·82 ,
a47·82 +1 := −2.
We recall that C can be increasing or decreasing only on
[a4j , b4j ],
[c4j , a4j+1 ],
j ∈ {1, . . . , 7 · 82 }.
We proceed further in the same way (as in the third and the fourth step). In the
2n-th step, we define continuous C so that
ε
kC(t) − C(t + 22n−1 )k < n+1 , t ∈ [−22n−1 − · · · − 2, −22n−3 − · · · − 2).
2
We get the nondecreasing sequence {d2n−1
} from
l
a2n−1
− 22n−1 ,
j
b2n−1
− 22n−1 ,
j
c2n−1
− 22n−1 ,
j
min{a2n−1
− 22n−1 + 82−2n /16, b2n−1
− 22n−1 },
j
j
max{a2n−1
− 22n−1 , b2n−1
− 22n−1 − 82−2n /16},
j
j
min{c2n−1
− 22n−1 , b2n−1
− 22n−1 + 82−2n /16},
j
j
max{c2n−1
− 22n−1 − 82−2n /16, b2n−1
− 22n−1 },
j
j
2n−1
min{c2n−1
− 22n−1 + 82−2n /16, a2n−1
},
j
j+1 − 2
2n−1
max{c2n−1
− 22n−1 , a2n−1
− 82−2n /16}
j
j+1 − 2
for j ∈ {1, . . . , 7 · 82n−3 }, from
c11 − 22n−1 ,
min{c11 − 22n−1 , b11 − 22n−1 + 82−2n /16},
max{c11 − 22n−1 − 82−2n /16, b11 − 22n−1 },
c12 − 22n−1 ,
min{c12 − 22n−1 , b12 − 22n−1 + 82−2n /16},
max{c12 − 22n−1 − 82−2n /16, b12 − 22n−1 },
min{a11 + (k − 1)(b11 − a11 )/(8 · 42n−3 ) − 22n−1 + 82−2n /16,
a11 + k(b11 − a11 )/(8 · 42n−3 ) − 22n−1 },
max{a11 + (k − 1)(b11 − a11 )/(8 · 42n−3 ) − 22n−1 ,
a11 + k(b11 − a11 )/(8 · 42n−3 ) − 22n−1 − 82−2n /16},
a11 + k(b11 − a11 )/(8 · 42n−3 ) − 22n−1 ,
k ∈ {1, . . . , 8 · 42n−3 },
min{c11 + (k − 1)(a12 − c11 )/(8 · 42n−3 ) − 22n−1 + 82−2n /16,
c11 + k(a12 − c11 )/(8 · 42n−3 ) − 22n−1 },
max{c11 + (k − 1)(a12 − c11 )/(8 · 42n−3 ) − 22n−1 ,
c11 + k(a12 − c11 )/(8 · 42n−3 ) − 22n−1 − 82−2n /16},
c11 + k(a12 − c11 )/(8 · 42n−3 ) − 22n−1 ,
k ∈ {1, . . . , 8 · 42n−3 },
min{a12 + (k − 1)(b12 − a12 )/(8 · 42n−3 ) − 22n−1 + 82−2n /16,
a12 + k(b12 − a12 )/(8 · 42n−3 ) − 22n−1 },
max{a12 + (k − 1)(b12 − a12 )/(8 · 42n−3 ) − 22n−1 ,
a12 + k(b12 − a12 )/(8 · 42n−3 ) − 22n−1 − 82−2n /16},
a12 + k(b12 − a12 )/(8 · 42n−3 ) − 22n−1 ,
k ∈ {1, . . . , 8 · 42n−3 },
16
M. VESELÝ
EJDE-2011/29
min{c12 + (k − 1)(a13 − c12 )/(8 · 42n−3 ) − 22n−1 + 82−2n /16,
c12 + k(a13 − c12 )/(8 · 42n−3 ) − 22n−1 },
max{c12 + (k − 1)(a13 − c12 )/(8 · 42n−3 ) − 22n−1 ,
c12 + k(a13 − c12 )/(8 · 42n−3 ) − 22n−1 − 82−2n /16},
c12 + k(a13 − c12 )/(8 · 42n−3 ) − 22n−1 ,
c21 − 22n−1 ,
k ∈ {1, . . . , 8 · 42n−3 },
min{c21 − 22n−1 , b21 − 22n−1 + 82−2n /16},
max{c21 − 22n−1 − 82−2n /16, b21 − 22n−1 },
...
c26 − 22n−1 ,
min{c26 − 22n−1 , b26 − 22n−1 + 82−2n /16},
max{c26 − 22n−1 − 82−2n /16, b26 − 22n−1 },
min{a21 + (k − 1)(b21 − a21 )/(8 · 42n−4 ) − 22n−1 + 82−2n /16,
a21 + k(b21 − a21 )/(8 · 42n−4 ) − 22n−1 },
max{a21 + (k − 1)(b21 − a21 )/(8 · 42n−4 ) − 22n−1 ,
a21 + k(b21 − a21 )/(8 · 42n−4 ) − 22n−1 − 82−2n /16},
a21 + k(b21 − a21 )/(8 · 42n−4 ) − 22n−1 ,
k ∈ {1, . . . , 8 · 42n−4 },
min{c21 + (k − 1)(a22 − c21 )/(8 · 42n−4 ) − 22n−1 + 82−2n /16,
c21 + k(a22 − c21 )/(8 · 42n−4 ) − 22n−1 },
max{c21 + (k − 1)(a22 − c21 )/(8 · 42n−4 ) − 22n−1 ,
c21 + k(a22 − c21 )/(8 · 42n−4 ) − 22n−1 − 82−2n /16},
c21 + k(a22 − c21 )/(8 · 42n−4 ) − 22n−1 ,
k ∈ {1, . . . , 8 · 42n−4 },
...
min{a26
a26
+ (k − 1)(b26 − a26 )/(8 · 42n−4 ) − 22n−1 + 82−2n /16,
+ k(b26 − a26 )/(8 · 42n−4 ) − 22n−1 },
max{a26 + (k − 1)(b26 − a26 )/(8 · 42n−4 ) − 22n−1 ,
a26 + k(b26 − a26 )/(8 · 42n−4 ) − 22n−1 − 82−2n /16},
a26 + k(b26 − a26 )/(8 · 42n−4 ) − 22n−1 , k ∈ {1, . . . , 8 · 42n−4 },
min{c26 + (k − 1)(a27 − c26 )/(8 · 42n−4 ) − 22n−1 + 82−2n /16,
c26 + k(a27 − c26 )/(8 · 42n−4 ) − 22n−1 },
max{c26 + (k − 1)(a27 − c26 )/(8 · 42n−4 ) − 22n−1 ,
c26 + k(a27 − c26 )/(8 · 42n−4 ) − 22n−1 − 82−2n /16},
c26 + k(a27 − c26 )/(8 · 42n−4 ) − 22n−1 , k ∈ {1, . . . , 8 · 42n−4 },
...
c2n−2
1
−
2
, min{c2n−2
1
2n−2
max{c1
− 22n−1
2n−1
− 22n−1 , b2n−2
− 22n−1 + 82−2n /16},
1
− 82−2n /16, b2n−2
− 22n−1 },
1
...
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
2n−1
c2n−2
,
7·82n−4 − 2
17
2n−1 2n−2
min{c2n−2
, b7·82n−4 − 22n−1 + 82−2n },
7·82n−4 − 2
2n−1
2n−1
max{c2n−2
− 82−2n /16, b2n−2
},
7·82n−4 − 2
7·82n−4 − 2
min{a2n−2
+ (k − 1)(b2n−2
− a2n−2
)/8 − 22n−1 + 82−2n /16,
1
1
1
a2n−2
+ k(b12n−2 − a2n−2
)/8 − 22n−1 },
1
1
max{a2n−2
+ (k − 1)(b2n−2
− a2n−2
)/8 − 22n−1 ,
1
1
1
a2n−2
+ k(b2n−2
− a2n−2
)/8 − 22n−1 − 82−2n /16},
1
1
1
a2n−2
+ k(b2n−2
− a2n−2
)/8 − 22n−1 ,
1
1
1
k ∈ {1, . . . , 8},
min{c2n−2
+ (k − 1)(a2n−2
− c2n−2
)/8 − 22n−1 + 82−2n /16,
1
2
1
c2n−2
+ k(a22n−2 − c2n−2
)/8 − 22n−1 },
1
1
max{c2n−2
+ (k − 1)(a2n−2
− c2n−2
)/8 − 22n−1 ,
1
2
1
c2n−2
+ k(a2n−2
− c2n−2
)/8 − 22n−1 − 82−2n /16},
1
2
1
c2n−2
+ k(a2n−2
− c2n−2
)/8 − 22n−1 , k ∈ {1, . . . , 8},
1
2
1
...
2n−2
2n−2
2n−1
min{a2n−2
+ 82−2n /16,
7·82n−4 + (k − 1)(b7·82n−4 − a7·82n−4 )/8 − 2
2n−2
2n−2
2n−1
a2n−2
},
7·82n−4 + k(b7·82n−4 − a7·82n−4 )/8 − 2
2n−2
2n−2
2n−1
max{a2n−2
,
7·82n−4 + (k − 1)(b7·82n−4 − a7·82n−4 )/8 − 2
2n−2
2n−2
2n−2
2n−1
2−2n
a7·82n−4 + k(b7·82n−4 − a7·82n−4 )/8 − 2
−8
/16},
2n−2
2n−2
2n−2
2n−1
a7·82n−4 + k(b7·82n−4 − a7·82n−4 )/8 − 2
, k ∈ {1, . . . , 8},
2n−2
2n−2
2n−2
min{c7·82n−4 + (k − 1)(a7·82n−4 +1 − c7·82n−4 )/8 − 22n−1 + 82−2n /16,
2n−2
2n−2
2n−1
c2n−2
},
7·82n−4 + k(a7·82n−4 +1 − c7·82n−4 )/8 − 2
2n−2
2n−2
2n−2
max{c7·82n−4 + (k − 1)(a7·82n−4 +1 − c7·82n−4 )/8 − 22n−1 ,
2n−2
2n−2
2n−1
c2n−2
− 82−2n /16},
7·82n−4 + k(a7·82n−4 +1 − c7·82n−4 )/8 − 2
2n−2
2n−2
2n−1
c2n−2
, k ∈ {1, . . . , 8},
7·82n−4 + k(a7·82n−4 +1 − c7·82n−4 )/8 − 2
and from a number of b11 − 22n−1 such that the total number of d2n−1
is 21 · 82n−2 .
l
We denote
2n−1
a2n
,
1 := d1
c2n−1
7·82n−2
:=
2n−1
b2n
,
1 := d2
d321·82n−2 ,
2n−1
c2n
,
1 := d3
a2n−1
7·82n−2 +1
2n−3
:= −2
...
− · · · − 2.
In the (2n + 1)-th step, we define continuous C so that
ε
kC(t) − C(t − 22n )k < n+2 , t ∈ (2 + · · · + 22n−2 , 2 + · · · + 22n ].
2
Now C has constant values on [b2n+1
, c2n+1
], j ∈ {1, . . . , 7 · 82n−1 }, where we put
j
j
a2n+1
:= 2 + 22 + · · · + 22n−2
1
and we obtain
b2n+1
,
1
c2n+1
,
1
a2n+1
,
2
...
c2n+1
7·82n−1 ,
a2n+1
7·82n−1 +1
from the nondecreasing sequence of
2n
a2n
j+1 + 2 ,
2n
b2n
j +2 ,
2n
c2n
j +2 ,
18
M. VESELÝ
EJDE-2011/29
2n
min{a2n
+ 81−2n /16, bj2n + 22n },
j +2
2n 2n
2n
max{a2n
− 81−2n /16},
j + 2 , bj + 2
2n 2n
2n
min{c2n
+ 81−2n /16},
j + 2 , bj + 2
2n
2n
max{c2n
− 81−2n /16, b2n
j +2
j + 2 },
2n
2n
min{c2n
+ 81−2n /16, a2n
j +2
j+1 + 2 },
2n 2n
2n
max{c2n
− 81−2n /16}
j + 2 , aj+1 + 2
for j ∈ {1, . . . , 7 · 82n−2 } and
c11 + 22n ,
min{c11 + 22n , b11 + 22n + 81−2n /16},
max{c11 + 22n − 81−2n /16, b11 + 22n },
c12 + 22n ,
min{c12 + 22n , b12 + 22n + 81−2n /16},
max{c12 + 22n − 81−2n /16, b12 + 22n },
min{a11 + (k − 1)(b11 − a11 )/(8 · 42n−2 ) + 22n + 81−2n /16,
a11 + k(b11 − a11 )/(8 · 42n−2 ) + 22n },
max{a11 + (k − 1)(b11 − a11 )/(8 · 42n−2 ) + 22n ,
a11 + k(b11 − a11 )/(8 · 42n−2 ) + 22n − 81−2n /16},
a11 + k(b11 − a11 )/(8 · 42n−2 ) + 22n ,
k ∈ {1, . . . , 8 · 42n−2 },
min{c11 + (k − 1)(a12 − c11 )/(8 · 42n−2 ) + 22n + 81−2n /16,
c11 + k(a12 − c11 )/(8 · 42n−2 ) + 22n },
max{c11 + (k − 1)(a12 − c11 )/(8 · 42n−2 ) + 22n ,
c11 + k(a12 − c11 )/(8 · 42n−2 ) + 22n − 81−2n /16},
c11 + k(a12 − c11 )/(8 · 42n−2 ) + 22n ,
k ∈ {1, . . . , 8 · 42n−2 },
min{a12 + (k − 1)(b12 − a12 )/(8 · 42n−2 ) + 22n + 81−2n /16,
a12 + k(b12 − a12 )/(8 · 42n−2 ) + 22n },
max{a12 + (k − 1)(b12 − a12 )/(8 · 42n−2 ) + 22n ,
a12 + k(b12 − a12 )/(8 · 42n−2 ) + 22n − 81−2n /16},
a12 + k(b12 − a12 )/(8 · 42n−2 ) + 22n ,
k ∈ {1, . . . , 8 · 42n−2 },
min{c12 + (k − 1)(a13 − c12 )/(8 · 42n−2 ) + 22n + 81−2n /16,
c12 + k(a13 − c12 )/(8 · 42n−2 ) + 22n },
max{c12 + (k − 1)(a13 − c12 )/(8 · 42n−2 ) + 22n ,
c12 + k(a13 − c12 )/(8 · 42n−2 ) + 22n − 81−2n /16},
c12 + k(a13 − c12 )/(8 · 42n−2 ) + 22n ,
c21 + 22n ,
k ∈ {1, . . . , 8 · 42n−2 },
min{c21 + 22n , b21 + 22n + 81−2n /16},
max{c21 + 22n − 81−2n /16, b21 + 22n },
...
c26
2n
+ 2 , min{c26 + 22n , b26 + 22n + 81−2n /16},
max{c26 + 22n − 81−2n /16, b26 + 22n },
min{a21 + (k − 1)(b21 − a21 )/(8 · 42n−3 ) + 22n + 81−2n /16,
a21 + k(b21 − a21 )/(8 · 42n−3 ) + 22n },
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
max{a21 + (k − 1)(b21 − a21 )/(8 · 42n−3 ) + 22n ,
a21 + k(b21 − a21 )/(8 · 42n−3 ) + 22n − 81−2n /16},
a21 + k(b21 − a21 )/(8 · 42n−3 ) + 22n ,
k ∈ {1, . . . , 8 · 42n−3 },
min{c21 + (k − 1)(a22 − c21 )/(8 · 42n−3 ) + 22n + 81−2n /16,
c21 + k(a22 − c21 )/(8 · 42n−3 ) + 22n },
max{c21 + (k − 1)(a22 − c21 )/(8 · 42n−3 ) + 22n ,
c21 + k(a22 − c21 )/(8 · 42n−3 ) + 22n − 81−2n /16},
c21 + k(a22 − c21 )/(8 · 42n−3 ) + 22n ,
k ∈ {1, . . . , 8 · 42n−3 },
...
min{a26
a26
+ (k − 1)(b26 − a26 )/(8 · 42n−3 ) + 22n + 81−2n /16,
+ k(b26 − a26 )/(8 · 42n−3 ) + 22n },
max{a26 + (k − 1)(b26 − a26 )/(8 · 42n−3 ) + 22n ,
a26 + k(b26 − a26 )/(8 · 42n−3 ) + 22n − 81−2n /16},
a26 + k(b26 − a26 )/(8 · 42n−3 ) + 22n , k ∈ {1, . . . , 8 · 42n−3 },
min{c26 + (k − 1)(a27 − c26 )/(8 · 42n−3 ) + 22n + 81−2n /16,
c26 + k(a27 − c26 )/(8 · 42n−3 ) + 22n },
max{c26 + (k − 1)(a27 − c26 )/(8 · 42n−3 ) + 22n ,
c26 + k(a27 − c26 )/(8 · 42n−3 ) + 22n − 81−2n /16},
c26 + k(a27 − c26 )/(8 · 42n−3 ) + 22n , k ∈ {1, . . . , 8 · 42n−3 },
...
c2n−1
1
+ 2 , min{c2n−1
+ 22n , b2n−1
+ 22n + 81−2n /16},
1
1
max{c2n−1
+ 22n − 81−2n /16, b2n−1
+ 22n },
1
1
2n
...
c2n−1
7·82n−3
2n 2n−1
2n
+ 2 , min{c2n−1
+ 81−2n /16},
7·82n−3 + 2 , b7·82n−3 + 2
2n
2n
max{c2n−1
− 81−2n /16, b2n−1
7·82n−3 + 2
7·82n−3 + 2 },
min{a2n−1
+ (k − 1)(b2n−1
− a2n−1
)/8 + 22n + 81−2n /16,
1
1
1
a2n−1
+ k(b2n−1
− a2n−1
)/8 + 22n },
1
1
1
max{a2n−1
+ (k − 1)(b2n−1
− a2n−1
)/8 + 22n ,
1
1
1
a2n−1
+ k(b2n−1
− a2n−1
)/8 + 22n − 81−2n /16},
1
1
1
a2n−1
+ k(b2n−1
− a2n−1
)/8 + 22n , k ∈ {1, . . . , 8},
1
1
1
min{c2n−1
+ (k − 1)(a2n−1
− c2n−1
)/8 + 22n + 81−2n /16,
1
2
1
c2n−1
+ k(a2n−1
− c2n−1
)/8 + 22n },
1
2
1
max{c2n−1
+ (k − 1)(a2n−1
− c2n−1
)/8 + 22n ,
1
2
1
c2n−1
+ k(a2n−1
− c2n−1
)/8 + 22n − 81−2n /16},
1
2
1
c2n−1
+ k(a2n−1
− c2n−1
)/8 + 22n , k ∈ {1, . . . , 8},
1
2
1
2n
19
20
M. VESELÝ
EJDE-2011/29
...
2n−1
2n−1
2n
min{a2n−1
+ 81−2n /16,
7·82n−3 + (k − 1)(b7·82n−3 − a7·82n−3 )/8 + 2
2n−1
2n−1
2n
a2n−1
7·82n−3 + k(b7·82n−3 − a7·82n−3 )/8 + 2 },
2n−1
2n−1
2n
max{a2n−1
7·82n−3 + (k − 1)(b7·82n−3 − a7·82n−3 )/8 + 2 ,
2n−1
2n−1
2n
a2n−1
− 81−2n /16},
7·82n−3 + k(b7·82n−3 − a7·82n−3 )/8 + 2
2n−1
2n−1
2n
a2n−1
k ∈ {1, . . . , 8},
7·82n−3 + k(b7·82n−3 − a7·82n−3 )/8 + 2 ,
2n−1
2n−1
2n−1
min{c7·82n−3 + (k − 1)(a7·82n−3 +1 − c7·82n−3 )/8 + 22n + 81−2n /16,
2n−1
2n−1
2n
c2n−1
7·82n−3 + k(a7·82n−3 +1 − c7·82n−3 )/8 + 2 },
2n−1
2n−1
2n
max{c2n−1
7·82n−3 + (k − 1)(a7·82n−3 +1 − c7·82n−3 )/8 + 2 ,
2n−1
2n−1
2n
c2n−1
− 81−2n /16},
7·82n−3 + k(a7·82n−3 +1 − c7·82n−3 )/8 + 2
2n−1
2n−1
2n
c2n−1
k ∈ {1, . . . , 8},
7·82n−3 + k(a7·82n−3 +1 − c7·82n−3 )/8 + 2 ,
and the corresponding number of b11 + 22n .
Using this construction, we obtain a continuous function C on R. From Theorem
3.1 it follows that C is almost periodic. Since
kC(t)k = 0,
kC(t) − C(t + 2)k < ε/4,
t ∈ [0, 1],
kC(t)k < ε/4,
t ∈ (1, 2],
t ∈ [−2, 0), kC(t) − C(t − 4)k < ε/8,
...
t ∈ (2, 6],
kC(t) − C(t + 22n−1 )k < ε/2n+1 ,
t ∈ [−22n−1 − · · · − 23 − 2, −22n−3 − · · · − 23 − 2),
kC(t) − C(t − 22n )k < ε/2n+2 ,
we see that
kC(t)k <
t ∈ (2 + 22 + · · · + 22n−2 , 2 + 22 + · · · + 22n ],
∞
X
2ε
= ε,
j+1
2
j=1
t ∈ R.
(4.2)
We denote
In := [2 + 22 + · · · + 22n−2 , 2 + 22 + · · · + 22n ].
We will prove that we can choose constant values of C(t), t ∈ In on subintervals
with the total length denoted by r2n+1 which is grater than 22n−1 for all n ∈ N.
We can choose C on
[4 − 2 + 1/16 + 8−1 /16, 4 − 2 + 1 − 1/16 − 8−1 /16] ⊂ [2, 6],
[4 − 2 + 1 + 1/4 + 1/16 + 8−1 /16, 4 − 2 + 1 + 3/4 − 1/16 − 8−1 /16] ⊂ [2, 6],
[4 + 8−1 /16, 4 + 1 − 8−1 /16], [4 + 1 + 1/4 + 8−1 /16, 4 + 1 + 3/4 − 8−1 /16] ⊂ [2, 6].
Hence,
r3 ≥ 55/64 + 23/64 + 63/64 + 31/64 = 43/16;
(4.3)
i.e., the statement is valid for n = 1. We use the induction principle with respect
to n. Assume that the statement is true for 1, 2, . . . , n−1 and prove it for n. Without
loss of the generality (consider the below given process), we can also assume that
the estimation r2j > 22(j−1) is valid for j ∈ {1, . . . , n} (note that r2 = 5/4 > 20 ) if
we use analogous notation.
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
21
In view of the construction, we see that we can choose C on any interval
[s + 22n + 81−2n /16, t + 22n − 81−2n /16]
if we can choose C on [s, t], where s = blj < clj = t, l < 2n + 1. Especially, we can
choose C on
[22n + 81−2n /16, 1 + 22n − 81−2n /16],
[1 + 1/4 + 22n + 81−2n /16, 1 + 3/4 + 22n − 81−2n /16],
[−2 + 1/16 + 22n + 81−2n /16, −2 + 1 − 1/16 + 22n − 81−2n /16],
[−2 + 1 + 1/4 + 1/16 + 22n + 81−2n /16, −2 + 1 + 3/4 − 1/16 + 22n − 81−2n /16]
and on less than 7 · 82n−1 − 4 subintervals of In . Expressing
In =[0 + 22n , 1 + 22n ] ∪ [1 + 22n , 2 + 22n ] ∪ [−2 + 22n , 0 + 22n ] ∪ . . .
∪ [2 + 22 + · · · + 22n−4 + 22n , 2 + 22 + · · · + 22n−2 + 22n ]
∪ [−2
2n−1
3
2n
2n−3
− · · · − 2 − 2 + 2 , −2
3
(4.4)
2n
− ··· − 2 − 2 + 2 ]
and using the induction hypothesis, the construction, and (4.3), we obtain that we
can choose C on intervals of the lengths grater than or equal to
1 − 2 · 81−2n /16,
1 − 1/8 − 2 · 81−2n /16,
1/2 − 2 · 81−2n /16,
1/2 − 1/8 − 2 · 81−2n /16,
43/16 + 22 + 23 + · · · + 22n−3 + 22n−2 − 2 · 81−2n /16 · (7 · 82n−1 − 4).
Summing, we obtain
1 7 3 11
7
+ + +
+ 22n−1 − 2 − > 22n−1 ,
2 8 8 16
8
which is the above statement. Analogously, we can prove
r2n+1 ≥ 1 +
r2n > 22n−2 ,
n ∈ N.
(4.5)
(4.6)
Now we describe the principal fundamental matrix XC on In for arbitrary n ∈ N.
Since C is constant a has the form diag[ia, ia, . . . , ia] for some a ∈ R on each interval
[b2n+1
, c2n+1
], j ∈ {1, . . . , 6 · 42n−1 }, from
j
j
Z t2
XC (t2 ) − XC (t1 ) =
C(τ )XC (τ ) dτ, t1 , t2 ∈ R,
t1
we obtain
kXC (t) − XC2n+1 (t)k ≤
Z
k Z b2n+1
X
j
kC(τ )XC (τ )kdτ +
j=1
if t ≤
a2n+1
k+1 ,
aj2n+1
a2n+1
j+1
c2n+1
j
kC(τ )XC (τ )kdτ
(4.7)
t ∈ In , where
XC2n+1 (t) := XC (2 + 22 + · · · + 22n−2 ),
t ∈ [2 + 22 + · · · + 22n−2 , b2n+1
],
1
XC2n+1 (t) := exp(C(b2n+1
)(t − b2n+1
)) · XC2n+1 (b2n+1
), t ∈ (b2n+1
, c2n+1
],
1
1
1
1
1
XC2n+1 (t) := XC 2n+1 (c2n+1
),
1
...
t ∈ (c2n+1
, b2n+1
],
1
2
22
M. VESELÝ
EJDE-2011/29
2n+1
2n+1 2n+1
XC2n+1 (t) := exp(C(b2n+1
(b7·82n−1 ),
7·82n−1 )(t − b7·82n−1 )) · XC
2n+1
t ∈ (b2n+1
7·82n−1 , c7·82n−1 ],
XC2n+1 (t) := XC2n+1 (c2n+1
7·82n−1 ),
2
2n
t ∈ (c2n+1
7·82n−1 , 2 + 2 + · · · + 2 ].
It is seen that XC is bounded (see also the below given, where it is shown that
XC (t) ∈ U (m) for all t) as almost periodic C. Any interval
[2 + · · · + 22n−2 + l − 1, 2 + · · · + 22n−2 + l],
l ∈ {1, . . . , 22n }, n ∈ N
contains at most 42n+1 subintervals where C can be linear. Indeed, it suffices to
consider the construction. We repeat that the length of each one of the considered
subintervals is 81−2n /16 which implies that the total length of them on
Jnl := [2 + 22 + · · · + 22n−2 , 2 + 22 + · · · + 22n−2 + 22n−l ],
l ∈ {1, . . . , n}
is less than 21−l . Thus (consider also (4.7)), there exists K ∈ R such that
K
, t ∈ Jnl , l ∈ {1, . . . , n}, n ∈ N.
2l
From the form diag[ia(t), . . . , ia(t)] of all matrices C(t), we see that
kXC (t) − XC2n+1 (t)k ≤
kC(t)k = |a(t)|,
(4.8)
t ∈ R.
For simplicity, let a(t) ≥ 0, t ∈ R. Let anj ∈ R, j ∈ {1, . . . , n} be arbitrarily chosen.
Considering the construction and combining (4.5) and (4.6), we obtain that we can
choose constant values of C(t), t ∈ [2 + · · · + 22n−2 + (l − 1)2n , 2 + · · · + 22n−2 + l 2n ]
on subintervals with the total length grater than 2n−2 for each l ∈ {1, . . . , 2n } and
all sufficiently large n ∈ N. Since we choose C only so that
kC(t) − C(t − 22n )k < ε/2n+2 ,
t ∈ In ,
we see that we can obtain
XC2n+1 (tnj ) = diag[exp(ianj ), . . . , exp(ianj )]
for arbitrary tnj such that
tn1 ≥ 2 + 22 + 24 + · · · + 22n−2 + 3n − 30 ,
...
tnn ≥ tnn−1 + 3n − 3n−1 ,
tn2 ≥ tn1 + 3n − 31 ,
2 + 22 + 24 + · · · + 22n ≥ tnn
(4.9)
because we have
4n > n(3n − 30 ) > 3n − 30 > · · · > 3n − 3n−1 > 22n−k+1
for sufficiently large n ∈ N and some k = k(n) ∈ {1, . . . , n} satisfying
22n−k−2 · ε · 2−n−2 > 2π.
We recall that we need to prove the existence of such C, given by the above construction, for which the vector valued function XA (t) · XC (t) · u, t ∈ R is not almost
periodic for any u ∈ Cm , kuk1 = 1. Since
∗
∗
∗
(XA (t)XA
(t))0 = A(t) XA (t) XA
(t) − XA (t) XA
(t) A(t),
0
t∈R
and since the constant function given by I is a solution of X = A · X − X · A,
X(0) = I, we have XA (t) ∈ U (m) for all t. Thus, XC (t) ∈ U (m), t ∈ R as well.
∗
We add that XA (t) · XA
(t) = I, t ∈ R implies A∗ (t) + A(t) = 0, t ∈ R.
Let c ∈ C, |c| = 1, and N ∈ U (m) be arbitrarily given. Obviously, for any
M ∈ U (m), we can choose a number a(M, c) ∈ [0, 2π) in order that all eigenvalues
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
23
of matrix P := M · diag[exp(ia(M, c)), . . . , exp(ia(M, c))] are not in the neighbourhood of c with a given radius which depends only on dimension m. Indeed, if
M has eigenvalues λ1 , . . . , λm , then the eigenvalues of P are λ1 exp(ia(M, c)), . . . ,
λm exp(ia(M, c)). Considering
M diag[exp(ia(M, c)), . . . , exp(ia(M, c))]u − N u
and expressing vectors u ∈ Cm , |u1 | + · · · + |um | = 1, as linear combinations of the
eigenvectors of P , we see that
M diag[exp(ia(M, c)), . . . , exp(ia(M, c))]u
cannot be in a neighbourhood of N · u for some c ∈ C, |c| = 1. Thus (the considered
multiplication of matrices and vectors is uniformly continuous), there exist ϑ > 0
and ξ > 0 such that, for any matrices M, N ∈ U (m), one can find a(M, N ) ∈ [0, 2π)
satisfying
kM diag[exp(iã), . . . , exp(iã)]u − N uk1 > ϑ,
(4.10)
u ∈ Cm , kuk1 = 1, ã ∈ (a(M, N ) − ξ, a(M, N ) + ξ).
We showed that we can construct C so that we obtain
XC2n+1 (tnj ) = diag[exp(ianj ), . . . , exp(ianj )]
for arbitrarily given anj ∈ [0, 2π) and any tnj satisfying (4.9) if n ∈ N is sufficiently
large and j ∈ {1, . . . , n}. Especially, for sufficiently large n ∈ N and for
tn1 := 2 + 22 + 24 + · · · + 22n−2 + 3n − 30 ,
tn2 := tn1 + 3n − 31 ,
...
(4.11)
tnn := tnn−1 + 3n − 3n−1 ,
we can choose all XC2n+1 (tnj ) in the form without any conditions. Hence, we obtain
diagonal matrices XC2n+1 (tnj ), j ∈ {1, . . . , n}, given by numbers
exp ia(XA (tnj ), XA (tnj − 3n + 3j−1 )XC (tnj − 3n + 3j−1 ))
on their diagonals.
It is seen from (4.11) that each
tnj ∈ [2 + 22 + · · · + 22n−2 , 2 + 22 + · · · + 22n−2 + n3n ].
Thus (see (4.8)), for any η > 0, we have
kXC (tnj ) − XC2n+1 (tnj )k < η
(4.12)
for sufficiently large n = n(η) ∈ N and j ∈ {1, . . . , n}. From (4.10) and (4.12) it
follows that
kXA (tnj ) XC (tnj )u − XA (tnj − 3n + 3j−1 )XC (tnj − 3n + 3j−1 )uk1 > ϑ
(4.13)
m
for u ∈ C , kuk1 = 1, sufficiently large n ∈ N, and j ∈ {1, . . . , n}.
By contradiction, suppose that there exists u ∈ Cm , kuk1 = 1, with the property
that XA (t) · XC (t) · u, t ∈ R is almost periodic. Applying Theorem 2.4 for
ψ(t) = XA (t)XC (t)u,
t ∈ R, sn = 3n , n ∈ N, ε = ϑ,
we obtain
kXA (t + 3n1 )XC (t + 3n1 )u − XA (t + 3n2 )XC (t + 3n2 )uk1 < ϑ,
t∈R
(4.14)
for all n1 , n2 from an infinite set N (ϑ) ⊆ N. If we rewrite (4.14) as
kXA (t)XC (t)u − XA (t + 3n2 − 3n1 )XC (t + 3n2 − 3n1 )uk1 < ϑ,
t ∈ R,
24
M. VESELÝ
EJDE-2011/29
then it is easy to see that (4.13) is not valid for infinitely many n ∈ N. This
contradiction proves the theorem.
References
[1] Amerio, L.; Prouse, G.: Almost-Periodic Functions and Functional Equations. Van Nostrand
Reinhold Company, New York, 1971.
[2] Bede, B.; Gal, S. G.: Almost periodic fuzzy-number-valued functions. Fuzzy Sets and Systems 147 (2004), no. 3, 385–403.
[3] Bochner, S.: Abstrakte fastperiodische Funktionen. Acta Math. 61 (1933), no. 1, 149–184.
[4] Bohr, H.: Zur Theorie der fastperiodischen Funktionen. I. Eine Verallgemeinerung der Theorie der Fourierreihen. Acta Math. 45 (1925), no. 1, 29–127.
[5] Bohr, H.: Zur Theorie der fastperiodischen Funktionen. II. Zusammenhang der fastperiodischen Funktionen mit Funktionen von unendlich vielen Variabeln; gleichmssige Approximation durch trigonometrische Summen. Acta Math. 46 (1925), no. 1-2, 101–214.
[6] Caraballo, T.; Cheban, D.: Almost periodic and almost automorphic solutions of linear differential/difference equations without Favard’s separation condition. I. J. Differential Equations 246 (2009), no. 1, 108–128.
[7] Coppel, W. A.: Almost periodic properties of ordinary differential equations. Ann. Mat. Pura
Appl. 76 (1967), no. 4, 27–49.
[8] Corduneanu, C.: Almost Periodic Functions. John Wiley and Sons, New York, 1968.
[9] Eliasson, L. H.: Reducibility and point spectrum for linear quasi-periodic skew-products. In:
Proceedings of the International Congress of Mathematicians, Vol. II (Berlin, 1998). Doc.
Math. 1998, Extra Vol. II, 779–787.
[10] Favard, J.: Sur certains systèmes différentiels scalaires linéaires et homogènes è coefficients
presque-périodiques. Ann. Mat. Pura Appl. 61 (1963), no. 4, 297–316.
[11] Favard, J.: Sur les équations différentielles linéaires àcoefficients presque-périodiques. Acta
Math. 51 (1927), no. 1, 31–81.
[12] Fink, A. M.: Almost Periodic Differential Equations. Lecture Notes in Math., vol. 377.
Springer-Verlag, Berlin, 1974.
[13] Funakosi, S.: On extension of almost periodic functions. Proc. Japan Acad. 43 (1967), 739–
742.
[14] Haraux, A.: A simple almost-periodicity criterion and applications. J. Differential Equations
66 (1987), no. 1, 51–61.
[15] Hu, Z.; Mingarelli, A. B.: Favard’s theorem for almost periodic processes on Banach space.
Dynam. Systems Appl. 14 (2005), no. 3-4, 615–631.
[16] Hu, Z.; Mingarelli, A. B.: On a theorem of Favard. Proc. Amer. Math. Soc. 132 (2004),
no. 2, 417–428.
[17] Ishii, H.: On the existence of almost periodic complete trajectories for contractive almost
periodic processes. J. Differential Equations 43 (1982), no. 1, 66–72.
[18] Jaiswal, A.; Sharma, P. L.; Singh, B.: A note on almost periodic functions. Math. Japon. 19
(1974), no. 3, 279–282.
[19] Janeczko, S.; Zajac, M.: Critical points of almost periodic functions. Bull. Polish Acad. Sci.
Math. 51 (2003), no. 1, 107–120.
[20] Johnson, R. A.: A linear almost periodic equation with an almost automorphic solution. Proc.
Amer. Math. Soc. 82 (1981), no. 2, 199–205.
[21] Kurzweil, J., Vencovská, A.: Linear differential equations with quasiperiodic coefficients.
Czechoslovak Math. J. 37(112) (1987), no. 3, 424–470.
[22] Kurzweil, J., Vencovská, A.: On a problem in the theory of linear differential equations with
quasiperiodic coefficients. In: Ninth international conference on nonlinear oscillations, Vol. 1
(Kiev, 1981), 214-217, 444. Naukova Dumka, Kiev, 1984.
[23] Lipnitskiı̆, A. ,V.: On the discontinuity of Lyapunov exponents of an almost periodic linear
differential system affinely dependent on a parameter. (Russian. Russian summary.) Differ.
Uravn. 44 (2008), no. 8, 1041–1049; translation in: Differ. Equ. 44 (2008), no. 8, 1072–1081.
[24] Lipnitskiı̆, A. V.: On the singular and higher characteristic exponents of an almost periodic
linear differential system that depends affinely on a parameter. (Russian. Russian summary.)
Differ. Uravn. 42 (2006), no. 3, 347–355, 430; translation in: Differ. Equ. 42 (2006), no. 3,
369–379.
EJDE-2011/29
CONSTRUCTION OF ALMOST PERIODIC FUNCTIONS
25
[25] Ortega, R.; Tarallo, M.: Almost periodic linear differential equations with non-separated
solutions. J. Funct. Anal. 237 (2006), no. 2, 402–426.
[26] Palmer, K. J.: On bounded solutions of almost periodic linear differential systems. J. Math.
Anal. Appl. 103 (1984), no. 1, 16–25.
[27] Sell, G. R.: A remark on an example of R. A. Johnson. Proc. Amer. Math. Soc. 82 (1981),
no. 2, 206–208.
[28] Tkachenko, V. I.: On linear almost periodic systems with bounded solutions. Bull. Austral.
Math. Soc. 55 (1997), no. 2, 177–184.
[29] Tkachenko, V. I.: On linear homogeneous almost periodic systems that satisfy the Favard
condition. (Ukrainian. English, Ukrainian summaries.) Ukraı̈n. Mat. Zh. 50 (1998), no. 3,
409–413; translation in: Ukrainian Math. J. 50 (1998), no. 3, 464–469.
[30] Tkachenko, V. I.: On linear systems with quasiperiodic coefficients and bounded solutions.
(Ukrainian. English, Ukrainian summaries.) Ukraı̈n. Mat. Zh. 48 (1996), no. 1, 109–115;
translation in: Ukrainian Math. J. 48 (1996), no. 1, 122–129.
[31] Tkachenko, V. I.: On reducibility of systems of linear differential equations with quasiperiodic skew-adjoint matrices. (Ukrainian. English, Ukrainian summaries.) Ukraı̈n. Mat. Zh. 54
(2002), no. 3, 419–424; translation in: Ukrainian Math. J. 54 (2002), no. 3, 519–526.
[32] Tkachenko, V. I.: On uniformly stable linear quasiperiodic systems. (Ukrainian. English,
Ukrainian summaries.) Ukraı̈n. Mat. Zh. 49 (1997), no. 7, 981–987; translation in: Ukrainian
Math. J. 49 (1997), no. 7, 1102–1108.
[33] Tornehave, H.: On almost periodic movements. Danske Vid. Selsk. Mat.-Fys. Medd. 28
(1954), no. 13, 1–42.
[34] Veselý, M.: Almost periodic sequences and functions with given values. To appear in Arch.
Math., 15 pp.
[35] Veselý, M.: Construction of almost periodic sequences with given properties. Electron. J.
Differential Equations 2008, no. 126, 1–22.
Michal Veselý
Department of Mathematics and Statistics, Masaryk University, Kotlářská 2, CZ-611
37 Brno, Czech Republic
E-mail address: michal.vesely@mail.muni.cz
Download